Satisfactory Alternate Recipes Guide: Hard Drives and Factory Tradeoffs

Alternate recipes are the biggest long-term power spike in Satisfactory, and they're all locked behind one item: the Hard Drive. There are 118 crash sites scattered around the map, and every single one holds exactly one drive. Feed those drives to the MAM and it offers you alternate recipes that build familiar parts from different ingredients. Sometimes they're cheaper on ore. Sometimes they run faster per machine. Sometimes they're just plain weird. What they're never, though, is a free upgrade. Every alternate is a trade, and picking the right trades for your factory is the whole game.

Finding Crash Sites

Crash sites are wrecked FICSIT freighters lying around the map. Each one has a drop-pod with the Hard Drive inside, plus a pile of loose parts sitting on the ground next to it. There are 118 of them, and that number matters later because it's more than you'll ever actually need.

The intended way to find them is the Object Scanner. Once you research Radio Signal Scanning in the MAM's quartz tree, the scanner gets a crash-site mode and starts pinging pods within range, beeping faster as you close in. Here's the catch: that research node costs 100 Crystal Oscillators, 100 Motors, and an Object Scanner. That's a mid-game shopping list, not an early one. So don't sit around waiting for the scanner to start drive hunting. Crash sites are big, obvious wrecks, and exploring along cliffs and ridgelines will find them on its own. The community interactive map also lists all 118 with their exact unlock costs and elevations, which saves a lot of sad wandering. Pay attention to altitude when you plan a trip, since some pods sit on top of spires and others are buried in caves.

Getting a pod open is its own little puzzle. There are three kinds:

• Free pods. The handle just works. Grab the drive and go.
• Part pods. A slot wants a specific part, and the part is consumed when you open the hatch. Rotors and Circuit Boards are the most common asks.
• Power pods. You have to feed the pod electricity. Most want less than 100 MW, only seven in the whole world want more, and the worst one wants 400 MW.

For power pods, the classic move is packing materials for a few Biomass Burners and wiring them up on the spot. Setting up coal or fuel power for a one-time job is a pain. If you hook the pod into your main grid instead, unplug it after, because the pod keeps drawing power even once the drive is gone. That's a sneaky way to brown out a base you didn't touch.

The loose parts scattered around each site are worth grabbing too. They're fixed, same loot in every save, and sites constantly hand you exactly the part a neighboring pod wants. Chain-looting two or three sites in one trip is normal.

Bring gear for the neighborhood, not just the pod:

| Threat | What handles it | | --- | --- | | Poison gas and spore flowers | Gas Mask (cheap, Mycelia tree) or Nobelisk | | Uranium deposits and waste | Hazmat Suit | | Alpha hogs, spitters, stingers | Rifle + ammo, or a Xeno-Basher | | Cliffs and tall spires | Jetpack, Zipline, or just foundations |

A pocket stack of common parts saves return trips. Across all 118 pods, the full shopping list runs deep into late-game items, but the big early ones look like this:

| Item | Total needed (all sites) | | --- | --- | | Steel Beam | 150 | | Electromagnetic Control Rod | 84 | | Silica | 70 | | Rubber | 70 | | Quickwire | 63 | | Plastic | 60 |

Once a drive is out, the pod and the wreck chunks around it can be dismantled for parts (that came in with version 1.1). You must grab the Hard Drive first, or the game won't let you tear the pod down.

Researching Hard Drives

The MAM is cheap to build: 5 Reinforced Iron Plates, 15 Cable, and 45 Wire, unlocked by the Tier 1 Field Research milestone. Drop a Hard Drive in and it starts a 10-minute scan. Only one drive can research at a time, no matter how many MAMs you own, and the timer keeps ticking even if you dismantle the machine. That leads to a trick worth stealing: carry MAM materials with you, build one at a crash site, feed it the drive, tear it down, and keep hiking. Your backpack full of drives slowly cooks itself while you explore.

When the scan finishes, you get a choice of two alternate recipes, not three like the old days. You can reroll once for a different pair, so one drive can show you up to four recipes. Picked recipes never show up again, and since version 1.0 there's a Hard Drive library in the MAM that stores pending choices. Two things follow from that:

• You can park a bad-looking pair and decide later, instead of being forced into a pick.
• Leaving unwanted options sitting in the library removes them from the pool, so future scans (and rerolls) get more likely to land what you actually want.

The pool of possible recipes is gated by your progress. Each recipe has prerequisites, like reaching Tier 3 or finishing a MAM node, and recipes you can't qualify for yet just don't appear. The pair is locked in the moment the scan starts, so unlocking a milestone mid-scan changes nothing. And if literally nothing in the pool is available to you, the MAM refuses the drive with a "no rewards available" message, which mostly happens very early or after you've cleaned the pool out.

Two of your scans aren't recipes at all. The library also offers two inventory expansions worth +6 pocket slots each, and the first one has no prerequisites at all. Six extra slots from your first-ever drive is a pretty nice consolation prize when the recipe gods are unkind.

Budget a few drives for the sulfur tree, too. Four MAM nodes there, Compacted Coal, Turbofuel, Rocket Fuel, and Ionized Fuel, each eat one Hard Drive as a research cost instead of scanning it. Researching Compacted Coal has a nice side effect: it unlocks the Biocoal and Charcoal alternates for free, which otherwise would cost you two scans. That's the rare case where the "right" answer is knowable in advance.

Run the math and the world is generous. Unlocking every recipe plus both pocket expansions plus the four sulfur nodes comes to 111 drives, and the map holds 118. Seven spare, before you touch the FICSMAS advent calendar bonus. Late game, after Tier 8's Particle Enrichment milestone, the AWESOME Shop sells drives for 100 coupons apiece, which is mostly a flex since you'll have spares long before you have coupons burning a hole in your pocket.

One more for the toolbox: alternates only run in machines. Constructors, assemblers, foundries, refineries. You can't hand-craft any of them at a craft bench, so there's no sneaky early-game workaround.

Evaluating Input Costs

The wiki says it plainly: alternate recipes cannot be considered upgrades. Every one of them shifts the cost somewhere. Your job is deciding which costs you care about. Four things to look at before committing a factory wing to a new recipe:

1. Ore per finished part. The base cost of everything.
2. Machine count. More machines means more power, more belts, more space.
3. Throughput per machine. Some alts triple output in the same building.
4. Weird inputs. Water, rubber, petroleum coke. Do you have them, or are you about to build a whole new supply chain?

A worked example beats theory here. Take the Reinforced Iron Plate, the part everyone's first factory chokes on.

| Recipe | Per craft | | --- | --- | | Default | 6 Iron Plate + 12 Screws → 1 Reinforced Iron Plate | | Stitched Iron Plate | 10 Iron Plate + 20 Wire → 3 Reinforced Iron Plates |

Count ingots. Default plates cost 9 Iron Ingots (3 ingots make 2 plates), and those 12 screws want 3 more ingots through the rod-and-screw chain. Call it 12 ingots per plate, plus a three-machine chain feeding the assembler. Stitched makes 3 plates from 10 plates + 20 wire. Per plate that's about 5 iron ingots and 3.3 copper ingots (1 copper ingot makes 2 wire). So the iron bill drops by more than half, the entire screw production line evaporates, and the price is a modest copper habit. If your copper nodes are busy elsewhere, that trade might stink. If they're idle, it's a steal.

The same lens on other common parts:

• Concrete. Default is 3 Limestone → 1 Concrete. Wet Concrete runs 6 Limestone + 5 Water → 4 Concrete in a refinery. Same rock buys twice the output; you're paying with a water extractor and pipes instead.
• Steel. Default foundry recipe is 3 Iron Ore + 3 Coal → 3 Steel Ingots, so 1 ore per ingot. Coke Steel Ingot runs 15 Iron Ore + 15 Petroleum Coke → 20 ingots. That's 0.75 ore per ingot, plus it swaps coal out for a refinery byproduct you might already be dumping. Big win if you've got oil running.
• Machine speed. Bolted Iron Plate makes 3 Reinforced Iron Plates per 12-second craft versus the default's 1 per 12 seconds. Triple the output in one assembler, paid for with 250 Screws a minute. Screws for speed is a real trade, not a gimmick.

Power matters more than people think early. A Constructor pulls 4 MW and an Assembler 15 MW, so deleting a screw chain doesn't just save space, it deletes every constructor in it plus the belts. Screw-heavy recipes are the usual suspects: a dozen constructors quietly sipping 4 MW each adds up fast.

The simplest rule that never fails: shop your surplus. Drowning in limestone? Wet Concrete and Fine Concrete love you. Caterium pile sitting untouched? Caterium Wire turns 1 ingot into 8 Wire and takes pressure off your copper. Heavy Oil Residue backing up? Coated Cable eats it. The best alternate recipe is usually the one that eats whatever you have too much of.

Recipes for Early Bottlenecks

Every new save hits the same walls in roughly the same order: screws, then concrete, then steel, then quickwire. Here are the alternates that actually move the needle, with their prerequisites so you know when they can start showing up.

| Recipe | Per craft | Available after | What it fixes | | --- | --- | --- | --- | | Cast Screw | 5 Iron Ingot → 20 Screws | Nothing | Kills the rod step | | Iron Wire | 5 Iron Ingot → 9 Wire | Nothing | Copper-free wire | | Stitched Iron Plate | 10 Iron Plate + 20 Wire → 3 RIPs | Tier 2 | Screwless plates | | Bolted Iron Plate | 18 Iron Plate + 50 Screws → 3 RIPs | Tier 2 | 3x plates per assembler | | Wet Concrete | 6 Limestone + 5 Water → 4 Concrete | Tier 3 (Coal Power) | Half the limestone | | Solid Steel Ingot | 2 Iron Ingot + 2 Coal → 3 Steel Ingots | Tier 3 | ~33% less ore per ingot | | Steeled Frame | 2 RIPs + 10 Steel Pipe → 3 Frames | Tier 3 | Frame screw crunch | | Caterium Wire | 1 Caterium Ingot → 8 Wire | MAM Caterium | Copper relief | | Coated Cable | 5 Wire + 2 Heavy Oil Residue → 9 Cable | Tier 5 | Copper-light cable | | Plastic AI Limiter | 30 Quickwire + 7 Plastic → 2 AI Limiters | Caterium tree + Tier 5 | Quickwire hunger |

The two with no prerequisites at all, Cast Screw and Iron Wire, can roll on your very first drive. Cast Screw is the quiet star: it makes screws straight from ingots in one constructor, skipping the rod stage entirely. Same ingot-to-screw ratio as the default chain, but one machine type fewer, and those constructors you don't build are 4 MW each you don't pay.

Stitched Iron Plate is the famous one, and it earns the hype. Pair it with Cast Screw or wire production and your reinforced plate line stops needing screws at all. Default Modular Frames want 3 Reinforced Iron Plates + 12 Iron Rods per craft, so plates are the wall, and Stitched cuts plate hunger hard. Once frames are the squeeze, Steeled Frame takes 2 plates + 10 Steel Pipe → 3 frames, dropping the plate cost per frame from 1.5 down to about 0.67. The bill moves to steel, which you'll have by then.

Concrete bottlenecks are boring and real: foundations eat everything. Wet Concrete doubles your limestone efficiency, but note it's a refinery recipe, so it needs water piped in. Fine Concrete (3 Silica + 12 Limestone → 10) is even leaner on rock if you've got quartz going.

Around Tier 5, quickwire becomes the Screaming Part. AI Limiters default to 20 Quickwire each, and high-speed connectors want 40. Plastic AI Limiter makes 2 limiters from 30 Quickwire + 7 Plastic, which works out to 15 quickwire per limiter and drops the copper sheet demand entirely. Fused Wire (4 Copper Ingot + 1 Caterium Ingot → 30 Wire) and Fused Quickwire do similar work for wire-hungry lines.

A picking strategy that holds up: take the recipe that unblocks what you're building this week. A shiny Tier 7 aluminum recipe does nothing for you at Tier 4 except feel nice in the library. The pool thins as you take recipes, so patience gets rewarded, and rerolls are there when a pair is all duds. Worst case, park the choice in the library and keep the drive working on the next one.

Rebuilding Lines Safely

Here's the fact that makes rebuilds painless: dismantling anything in Satisfactory refunds 100% of its build cost. Every machine, every belt, every spluttering old coal generator. The only thing a rebuild ever really costs you is time, so experiment freely.

The safe pattern is build-beside, verify, then cut over:

1. Leave the old line running. Don't touch it yet.
2. Build the new recipe's line next to it (above or below works great on multi-floor factories).
3. Split some input off the existing source belts into the new line.
4. Watch the output belt's readout. The belt UI shows items per minute, so compare it against the numbers you planned.
5. Let the new line run a while and watch for starvation, where input rates don't quite keep up and machines sit idle.
6. Once it's stable, point the consumers at the new line, then dismantle the old one for a full refund.

That order matters because a half-torn-up factory produces nothing while you rebuild. Parallel lines keep the parts flowing the whole time, which your Space Elevator project will appreciate.

Blueprints make swaps much cheaper on effort. The Blueprint Designer Mk.1 unlocks at Tier 4 in the FICSIT Blueprints milestone (it costs 15 Modular Frames, 25 Cable, 100 Concrete, and 100 Steel Beams to build) and gives you a 32-meter cube to design in. Lay out a full tile of, say, six Cast Screw constructors with manifold belts and mergers, save it once, and stamp it everywhere. Mk.2 and Mk.3 designers at Tier 6 and Tier 9 just make the cube bigger.

Two smaller helpers worth using. The Dimensional Depot (one Mercer Sphere in the alien tech tree) lets machines upload to a central pool you can pull from anywhere, which means your rebuild crew never runs back to storage for belts. And remember that machines switch recipes in place. A constructor already making screws can just become a Cast Screw machine. Belts and manifolds usually survive a recipe swap fine; the ratios never do, which is why you check the belt readout instead of trusting the old layout.

So the loop closes nicely: drive in the MAM, recipe picked against your actual bottlenecks, new line proven beside the old one, then the old one sold back for every iron plate it cost. Pick the recipe first, rebuild second, and check that belt readout before you tear out anything that's currently feeding your milestones.